[WS:] Thanks for the link. I downloaded the entire database for 186 countries and run a quick correlation test between the data points that you quote - I got correlation coefficient 0.25 - which is positive but rather weak, meaning no clear correlation. Obviously, more factors affect homicide rates than just availability of guns so it would be interesting to run a model with controls for crime rates, which obviously would be a better predictor than gun ownership.
So let's conclude this discussion with the following point. There is no strong correlation between gun ownership and number of homicides based on the data that you provided. But since homicide rates are affected by so many socio-demographic factors the effect of gun ownership, if any at all, is not likely to be that great. To translate that from probability speak to real numbers - reducing the number of guns by thousands may save at best a dozen or so lives.
So at this point one can conclude that the outcome is not worth the effort and do nothing or one can conclude that saving this dozen or so lives is worth sacrificing thousands or even millions of guns. I opt for the latter. I see no use at all in having a gun, so I have no problems scrapping the constitutional right to own guns. This is all what I have to say on the subject. Thanks for your input and thanks for keeping it civil. Over and out.
-- Wojtek
"An anarchist is a neoliberal without money."